MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
arXiv:2308.00352
Abstract
Remarkable progress has been made on automated problem solving through societies of agents based on large language models (LLMs). Existing LLM-based multi-agent systems can already solve simple dialogue tasks. Solutions to more complex tasks, however, are complicated through logic inconsistencies due to cascading hallucinations caused by naively chaining LLMs. Here we introduce MetaGPT, an innovative meta-programming framework incorporating efficient human workflows into LLM-based multi-agent collaborations. MetaGPT encodes Standardized Operating Procedures (SOPs) into prompt sequences for more streamlined workflows, thus allowing agents with human-like domain expertise to verify intermediate results and reduce errors. MetaGPT utilizes an assembly line paradigm to assign diverse roles to various agents, efficiently breaking down complex tasks into subtasks involving many agents working together. On collaborative software engineering benchmarks, MetaGPT generates more coherent solutions than previous chat-based multi-agent systems. Our project can be found at https://github.com/geekan/MetaGPT
Cited by in corpus (11)
- AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges
- General Purpose Artificial Intelligence Systems (GPAIS): Properties, Definition, Taxonomy, Societal Implications and Responsible Governance
- Change-Agent: Towards Interactive Comprehensive Remote Sensing Change Interpretation and Analysis
- Integrating Graphs with Large Language Models: Methods and Prospects
- LAMBDA: A Large Model Based Data Agent
- SPROUT: an Interactive Authoring Tool for Generating Programming Tutorials with the Visualization of Large Language Models
- The Emerged Security and Privacy of LLM Agent: A Survey with Case Studies
- DesignRepair: Dual-Stream Design Guideline-Aware Frontend Repair with Large Language Models
- Flows: Building Blocks of Reasoning and Collaborating AI
- Predicting potentially abusive clauses in Chilean terms of services with natural language processing
- Agentic Social Affordance Framework (ASAF): Agent Identity Design as a Collaboration Interface in Multi-Agent Systems